Confocal Raman Microscope RAMANforce
Outstanding software is required to analyze the large amount of data from high quality Raman images. The software provided by Nanophoton is equipped with high-speed data processing capacity and miscellaneous analytical functions to support imaging analyses performed by RAMANforce. Parts of its functions are introduced here.
- Measurement Functions
- Analysis Functions
Fast, high-resolution 3D Raman Imaging
Using confocal optics, which detects Raman light inside the sample nondestructively, a 3D Raman image of a transparent sample can be obtained. This function repeats ultra-fast XY Raman imaging using line illumination several times while changing the stage height, and stacks the slices to create a 3-dimensional image in memory. By using the high imaging speed and depth resolution of RAMANforce, a more intuitive understanding can be obtained of the internal structure and component distribution inside a sample.
Brand-new wide field-of-view imaging along curved surface
RAMANforce has a brand-new wide field-of-view (FOV) Raman imaging capability. It automatically measures the surface height of a sample while capturing a wide FOV microscopic image, and makes it possible to measure a wide FOV Raman image with auto-focusing. As the images show on the right, a fine focused Raman image can be obtained from all over the surface of a curved tablet.
The measurement algorithm is further improved to speed up the imaging 3 times faster than our previous RAMANtouch model.
Fully automated particle scanning (option)
This function detects particles in a microscopic image, and Raman measurements are automatically carried out with quick auto-focus. Fully utilizing our fast and accurate laser beam scanning technology, Nanophoton has developed three different measurement modes: "auto point measurement at center of each particle", "auto scanning of whole surface of each particle to get averaged spectrum", and "auto Raman mapping of each particle with arbitrary scanning pitch". Detected spectra are automatically identified using a real-time spectrum search during measuring. Measuring and analysis according to size classes as defined by ISO 16232 is also possible.
Interlaced Raman Imaging for quick overview
Interlaced imaging scans a sample skipping some pixels to quickly obtain a rough Raman image, then keeps scanning the rest while the Raman image becomes finer and finer. This mode will be the best scanning mode for a quick overview of component distribution.
Real-time multivariate analysis Multivariate analysis is carried out during Raman imaging to show component spectra and their distribution in real time. Random scanning (Option) Sequential order of pixel measurement is determined randomly for each imaging mode to avoid sample heating by laser irradiation at a specific local area.
Quantitative analysis of components dispersiveness
This function is to obtain a quantitative analysis of component dispersiveness such as uniformity, aggregational states and locality in a Raman image. Such can be investigated from every possible angle by counting the number of particle outlines along an evaluation axis such as (X and Y) or (r and θ), or calculating the standard deviation of number of particles in a (grid) or (Voronoi diagram) layout.
Composition rate evaluation by area ratio analysis
The composition rate of a sample is evaluated from a Raman image which shows the distribution of components. It is calculated from a binary Raman image, which is made assuming that each pixel corresponds to only one component. Composition rate can also be calculated using CLS as described below.
Distribution and ratio evaluation of components present on the tablet surface
Raman Image and Analysis Result
■：Active pharmaceutical ingredient (API)
Result of Area Ratio Analysis
Particle size analysis of a component from the Raman image
Statistical analysis of particle sizes for a specific component can be carried out from the Raman image. Approximating a particle as an oval, size distribution can be indicated in a histogram on a number basis, area basis and volume basis. The statistics (maximum, minimum, average value, standard deviation, etc.) can also be calculated.
Particle size analysis of API in a tablet
Binary Image of API
Particle size analysis of API
※The size related information was deleted because samples are commercial products
Noise reduction by Singular Value Decomposition (SVD)
First extract the orthogonal spectra from the Raman image and place them in descending order of contribution. By reconstructing the Raman image after the removal of spectra with a low contribution ratio (i.e., noise reduction), a clear Raman image consisting of high S/N spectra can be obtained even if the S/N of raw data is non-optimal.
Noise reduction processing of Raman image of HeLa cells
A part of the extracted spectra by SVD
Linear combination analysis by Classical Least Squares (CLS)
Represent the unknown spectrum by a linear combination of known raw material spectra and calculate intensity of the known spectra by the least squares method. A quantitative evaluation of the composition is applicable using raw material spectra obtained under the same measurement conditions.
CLS analysis of cathode materials of a lithium-ion battery
Raman image calculated by CLS
■：Lithium cobalt oxide
Composition rate calculated by CLS
Raw material spectra used for calculation
Comparison of raw spectrum and CLS spectrum
Multivariate Curve Resolution (MCR)
MCR extrapolates the component spectra by assuming that the unknown spectrum can be expressed by linearly combining a finite number (N) of component spectra and that both the spectrum intensity and concentration of each component have non-negative values. ALS (Alternating Least Squares) (which allows for a quick calculation), and MUR (Multiplicative Update Rule) (which guarantees convergence) are both installed and available.
Raman imaging analysis of HeLa cells using MCR
Component spectra calculated by MCR
Raman image calculated by MCR
■：Component 3（Cytochrome c）